- From: Daniel Veillard <Daniel.Veillard@w3.org>
- Date: Sun, 12 Nov 2000 12:02:38 +0100
- To: Jason Diamond <jason@injektilo.org>
- Cc: www-rdf-interest <www-rdf-interest@w3.org>
On Sat, Nov 11, 2000 at 12:52:29PM -0800, Jason Diamond wrote: > Hello. > > > Push vs pull is tricky to do both but have been done on systems that > > are widely installed (on Linux-en) - see libpng for handling the PNG > > image format and Daniel Veillard's libxml that does both for XML. > > I haven't used it but looking at the documentation, I wouldn't consider > libxml to support a pull-based parser. Its version of pull loads the > document into a DOM from which you can then pull elements from. Well you don't have to build the tree. there is an expat SAX compatible interface, and DOM construction is done on top of it. The pull vs. push libxml interfaces may have a different meaning than the one you expect. In the pull interface, the parser keeps the control of the application loop, in the push interface the parser input is provided by the application as it receives the data. libxml pull interface is not asking the parser to return only selected nodes to the application. This can certainly be done on top of the SAX API but not something pertaining at the parser level IMHO. Daniel -- Daniel.Veillard@w3.org | W3C, INRIA Rhone-Alpes | libxml Gnome XML toolkit Tel : +33 476 615 257 | 655, avenue de l'Europe | http://xmlsoft.org/ Fax : +33 476 615 207 | 38330 Montbonnot FRANCE | Rpmfind search site http://www.w3.org/People/all#veillard%40w3.org | http://rpmfind.net/
Received on Sunday, 12 November 2000 06:02:43 UTC